The Underspend

What this pack runs.

A year of maintenance not done arrives in that year's operating profit. It is visible, it is attributable, and somebody is usually thanked for it. The work stays where it was, and nothing in the accounts records that half: no line called what the base did not get, no balance carrying it forward, no year reporting it as owing. Depreciation runs on historic cost and says nothing about condition. This pack turns it into a subtraction and a multiplication. Replacement value times the rate the base has to be held at gives what a year needs; less what is actually spent, times the years the gap has run, gives what has accumulated. Then the honest other side: the money not spent went into profit and has been earning the company's return ever since, so the accumulation is read against what holding it back was worth rather than against nothing. What decides it is the price of doing the work late — work that arrives as a failure costs a multiple of the same work scheduled — so the whole decision reduces to the break-even unplanned share, how much of the accumulated work can arrive as a failure before holding the money back has cost more than it earned. It stops on six conditions: no replacement value, no rate, no split between holding and growing, no start year for the gap, no measured multiple, and no view of what would arrive unplanned.

The casebooks these cases come from.

The pack reads nine sourced decisions through the framework. They are drawn from these casebooks, each one a set of verified records on a single decision type.

A verified record has passed a mechanical claim-versus-source check, a second model's read, and human approval. The count is what exists today, not a target.